HPSA and MUA/P Geographic Compilation Logic

The following information applies ONLY to the spatial data in the HRSA Data Warehouse Map Tool, geographic advisors, and feature service; the Health Professional Shortage Area (HPSA) and Medically Underserved Area/Population (MUA/P) data available in the Report Tool, MUA/P Find, and HPSA Find represent ALL of the data obtained from the respective source systems.

More about Health Profession Shortage Areas (HPSAs) and Medically Underserved Areas/Populations (MUAs/Ps).

The HRSA Data Warehouse creates the spatial data for the MUA/P and HPSA layers using a multi-step process. Varying types of geographic boundary (county, county subdivision, and Census tract) are combined into a single logical unit.

Each HPSA and MUA/P that consists of one or more geographic areas (components) contains geographic identifiers for each component. These identifiers correspond to the Federal Information Processing Standards (FIPS) codes for counties, and to the Census-assigned codes for county subdivision (also known as minor civil division, or MCD) and Census tract identifier.

In preparation for each decennial census of the U.S., the U.S. Census Bureau reviews the MCD and tract boundaries and makes adjustments as deemed necessary to support that particular census. In addition, as changes occur to the list of counties and/or the areas defined by their boundaries, the FIPS county list is adjusted. Thus, at least every ten years, some or all of the geographic identifiers are rendered obsolete, new identifiers are created, and county boundaries are changed.

The HPSA and MUA/P source databases in HRSA are not immediately updated to reflect changes in these identifiers. However, once the new identifiers are available, newly-designated areas use the most current set of identifiers. The existing data are updated at a later date, if at all. For HPSAs or MUA/Ps which have been withdrawn (are no longer designated) the identifiers may never be updated. Currently-designated areas are updated on an as-needed basis. Thus, for some period of time, the geographic identifiers in the source systems (and consequently in the HRSA Data Warehouse) represent a mix of data from more than one time period.

Prior to April 2005, the vast majority of geographic identifiers in both the HPSA and MUA/P databases were Census 1990 vintage. The corresponding county, MCD, and Census tract spatial layers were used as the basis of the spatial processing. Components which used unmatched identifiers (i.e., those from the 2000 Census) simply were not included in the final HPSA and MUA/P layers. The number of these components was small enough that the resulting inaccuracies were deemed by HRSA to be negligible for the purposes of the HRSA Data Warehouse Map Tool, geographic advisors, and feature service.

However, between the February 2005 and April 2005 HRSA Data Warehouse data refresh cycles, the geographic identifiers for the majority of designated MUA/Ps were updated from the 1990 to the 2000 Census geographies. In addition, a significant number of new or revised HPSA designations were also defined using the year 2000 identifiers. Thus, significant errors in the final spatial data would have been introduced had the building process not been altered.

Beginning with the July 2012 data release, the HPSA spatial data are built according to the following logic:

• All geographic identifiers are referring to Census 2010 geographic areas and assumed to match a county FIPS code, MCD identifier, or census tract identifier associated with the 2010 Census.
• Any identifiers which do not match Census 2010 identifiers are assumed to be erroneous. The geographic components which are associated with these identifiers are not included in the spatial data.

HPSA "Points" (Facilities):

In addition to HPSAs based on geography, some HPSAs are designated based on population concentrations such as correctional institutions or hospitals. These are incorporated into the HRSA Data Warehouse Map Tool, geographic advisors, and feature service through the process of geocoding, or correlating a street address with a particular point on the surface of the earth. If no address data are included in the source data it is not possible to make that correlation for a given designation record, and no point is created for display on a map which would represent that facility.

Beginning with the November 2012 data release, the MUA/P spatial data are built according to the following logic:

• All geographic identifiers are assumed to be Census 2010, if they match a county FIPS code, MCD identifier, or census tract identifier associated with the 2010 Census.